Trout Fishing in the U.S. Addendum to the 1996 National Survey of Fishing, Hunting and Wildlife-Associated Recreation Report 96-4

              Trout Fishing
in the U.S.
Addendum to the 1996 National
Survey of Fishing, Hunting and
Wildlife-Associated Recreation
U.S. Fish & Wildlife Service
Report 96-4
Trout Fishing
in the U.S.
Addendum to the 1996 National
Survey of Fishing, Hunting and
Wildlife-Associated Recreation
Report 96-4
May 1999
Genevieve Pullis
Division of Federal Aid
U.S. Fish and Wildlife Service
Washington, D.C.
and
Andrew Laughland
Division of Economics
U.S. Fish and Wildlife Service
Washington, D.C.
Division of Federal Aid
U.S. Fish and Wildlife Service
Washington, D.C. 20240
Director, Jamie Clark
Chief, Division of Federal Aid, Bob Lange
This report is intended to complement the National and State reports from the
1996 National Survey of Fishing, Hunting, and Wildlife-Associated Recreation.
The conclusions are the authors and do not represent official positions of the
U.S. Fish and Wildlife Service.
Front Cover—© PhotoDisc
U.S. Fish & Wildlife Service
Trout fishing is one of the most popular
types of fishing in the United States. In
1996, 31 percent of all freshwater anglers
in the United States fished for trout. This
report uses information from the 1996
National Survey of Fishing, Hunting and
Wildlife-Associated Recreation
(FHWAR) to describe trout fishing in the
United States and the characteristics of
trout anglers. For this report a trout
angler is anyone 16 years of age and older
who fished for trout at least once during
the year. They may have fished for other
species too. In this report, trout refers to
all trout, including rainbow, brown,
brook, lake, and so forth. Great Lakes
trout fishing raises some different issues
so it is not included in this report.
This report has five sections. The first
section deals with the extent of
participation in trout fishing. It presents
the number of trout anglers and fishing
days for the nation and by state. The
second section discusses the
demographic characteristics of trout
anglers and compares them to other
freshwater anglers and the overall U.S.
population. The third section considers
changes in trout fishing participation
through time by comparing results for
the 1991 and 1996 surveys. The fourth
section develops a statistical model of
trout fishing participation which predicts
whether or not a freshwater angler will
fish for trout or not, based on the
person’s demographic characteristics and
residency. The final section summarizes
the report.
2
Introduction
USFWS photo
As one of the most popular sport fish in
the United States in 1996, trout were
sought by 9 million freshwater anglers,
31 percent of all freshwater anglers
(Table 1). In comparison, 12.7 million
anglers (44 percent) fished for black
bass, 8 million (28 percent) fished for
panfish, 7.4 million (26 percent) fished
for catfish and bullheads, and 6.4 million
(22 percent) fished for crappie. Since
anglers can fish for more than one
species, the sum of the number of
anglers by species is larger than the
total number of anglers.
Anglers fished for trout on 94 million
days for an average of 10 days per angler.
Keeping in mind that anglers can fish
for more than one species in a day, trout
were sought on 19 percent of all
freshwater fishing days.
Tables 2, 3, and 4 contain state-by-state
estimates of trout fishing. These tables
present fishing estimates in each state by
residents and nonresidents combined.
The percent of freshwater anglers that
fish for trout varied widely by state —
from less than 6 percent in low
participation states such as Alabama,
Kentucky, Louisiana, Minnesota, and
South Carolina to greater than 80 percent
in Colorado, Idaho, Montana, Utah,
Washington and Wyoming (Table 2).
Generally, the Northeastern and Western
states had the highest levels of
participation in trout fishing. Southern
and Midwestern states had the lowest
levels of participation. An exception to
this is West Virginia which has a
relatively high participation rate (54
percent). This can be seen graphically in
Figure 1 which shows the percent of
anglers who fished for trout by state.
3
Trout Fishing
Participation Levels
Table 1. Freshwater Anglers and Days of Fishing by Type of Fish: 1996
(Population 16 years of age and older. Numbers in thousands.
Excludes Great Lakes fishing.)
Anglers Days of Fishing Average Days per
Type of fish Number Percent Number Percent Angler
Total, all types of fish 28,921 100 485,474 100 17
Black bass (largemouth, 12,708 44 191,350 39 15
smallmouth, etc.)
Trout 8,974 31 93,566 19 10
Panfish 7,961 28 103,041 21 13
Catfish and bullheads 7,430 26 91,498 19 12
Crappie 6,363 22 91,031 19 14
White bass, striped bass 4,756 16 61,386 13 13
and striped bass hybrids
Anything 4,218 15 39,035 8 9
Another type of freshwater fish 3,729 13 44,401 9 12
Walleye and sauger 3,276 11 48,726 10 15
Salmon 1,218 4 11,742 2 10
Steelhead 470 2 6,699 1 14
Note: Detail does not add to total because of multiple responses. Respondent identified “anything” from a
list of categories of fish.
Columns one and two of Table 3 show
the number of days spent freshwater
fishing and fishing for trout. The third
column shows the share of all freshwater
days that anglers spent fishing for trout.
These days do not represent fishing for
trout exclusively; the anglers could have
sought more than one species of fish on a
day of fishing. The pattern of days trout
fishing follows that of trout participation;
anglers in Northeastern and Western
states spend more days trout fishing
than anglers in Southern and
Midwestern states.
Table 4 shows the average number of
days of freshwater and trout fishing by
state. Nationally, anglers averaged 17
days of freshwater fishing and 10 days of
trout fishing. States with the highest
average days of trout fishing tended to
be states in the Northeast where trout
fishing is popular, such as Connecticut,
New Jersey, Massachusetts and
Rhode Island.
The average number of days anglers
spent fishing for trout, by state, is shown
graphically in Figure 2. Several states
have no marking on them. The FHWAR
Survey’s sample sizes for these states
were less than 10 so their average days
are not judged reliable enough to include
in the figure. The heavily shaded areas
show states where the average number
of trout fishing days is greater than
or equal to 11. Moderate shading
represents average days greater than
or equal to 8 and less than or equal to 10.
The lightly shaded states are states
where the average days are less than
or equal to 7.
4
Table 2. Freshwater and Trout Anglers, by
State Where Fishing Occurred: 1996
(Population 16 years of age and older.
Numbers in thousands. Excludes Great
Lakes fishing.)
Freshwater
Trout Anglers
State Anglers Number Percent
United States 28,921 8,974 31
Alabama 843 *30 *4
Alaska 313 111 35
Arizona 483 218 45
Arkansas 739 152 21
California 2,175 1,525 70
Colorado 787 699 89
Connecticut 318 168 53
Delaware 66 *9 *14
Florida 1,137 ** **
Georgia 967 160 17
Hawaii 22 *6 *27
Idaho 474 409 86
Illinois 1,123 74 7
Indiana 863 ** **
Iowa 477 *48 *10
Kansas 341 ** **
Kentucky 772 *39 *5
Louisiana 815 *39 *5
Maine 290 185 64
Maryland 319 89 28
Massachusetts 377 179 47
Michigan 1,311 204 16
Minnesota 1,421 *72 *5
Mississippi 487 ** **
Missouri 1,138 255 22
Montana 329 266 81
Nebraska 247 27 11
Nevada 219 159 73
New Hampshire 237 131 55
New Jersey 428 195 46
New Mexico 312 237 76
New York 1,111 468 42
North Carolina 1,009 197 20
North Dakota 90 *6 *7
Ohio 908 *64 *7
Oklahoma 891 ** **
Oregon 589 395 67
Pennsylvania 1,277 745 58
Rhode Island 72 39 54
South Carolina 716 *38 *5
South Dakota 213 42 20
Tennessee 767 120 16
Texas 2,147 *141 *7
Utah 397 341 86
Vermont 176 107 61
Virginia 761 239 31
Washington 768 628 82
West Virginia 323 174 54
Wisconsin 1,232 *77 *6
Wyoming 379 357 94
*Estimate based on small sample size.
**Sample size too small to report data reliably.
Table 3. Days Freshwater and Trout
Fishing, by State Where Fishing
Occurred: 1996
(Excludes Great Lakes fishing.)
Days of
Days Fishing
Trout Fishing
State in Freshwater Number Percent
United States 485,474 93,566 19
Alabama 14,256 *101 *1
Alaska 3,602 1,151 32
Arizona 4,689 1,579 34
Arkansas 9,661 635 7
California 28,987 16,291 56
Colorado 8,232 6,811 83
Connecticut 3,880 2,101 54
Delaware 980 *114 *12
Florida 18,409 ** **
Georgia 12,857 1,605 12
Hawaii 189 *26 *14
Idaho 4,411 3,324 75
Illinois 17,089 422 2
Indiana 13,456 ** **
Iowa 7,062 *490 *7
Kansas 6,355 ** **
Kentucky 9,631 *413 *4
Louisiana 18,493 *259 *1
Maine 4,107 2,149 52
Maryland 4,290 967 23
Massachusetts 6,746 2,891 43
Michigan 19,456 1,102 6
Minnesota 25,897 *465 *2
Mississippi 8,213 ** **
Missouri 14,682 2,052 14
Montana 2,617 2,100 80
Nebraska 3,004 219 7
Nevada 1,976 1,162 59
New Hampshire 3,139 1,412 45
New Jersey 6,021 2,476 41
New Mexico 2,836 2,046 72
New York 17,412 3,161 18
North Carolina 15,831 1,906 12
North Dakota 1,321 *87 *7
Ohio 12,878 *220 *2
Oklahoma 14,674 ** **
Oregon 7,117 3,524 50
Pennsylvania 18,635 8,861 48
Rhode Island 1,347 683 51
South Carolina 11,341 *150 *1
South Dakota 2,748 380 14
Tennessee 11,317 1,083 10
Texas 37,575 *503 *1
Utah 3,936 3,084 78
Vermont 1,951 888 46
Virginia 9,282 1,931 21
Washington 10,975 7,168 65
West Virginia 5,040 1,881 37
Wisconsin 14,398 *658 *5
Wyoming 2,415 2,118 88
*Estimate based on small sample size.
**Sample size too small to report data reliably.
5
Figure 1. Percent of Freshwater Anglers Who Sought Trout
Small sample size
≤ 15 percent
16–49 percent
≥ 50 percent
Figure 2. Average Number of Trout Fishing Days
Small sample size
≤ 7 days
8–10 days
≥ 11 days
Table 4. Average Days Freshwater and
Trout Fishing, by State Where Fishing
Occurred: 1996
(Excludes Great Lakes fishing.)
Average Days Average Days
State Freshwater Trout
United States 16.7 10.5
Alabama 16.9 *3.4
Alaska 11.5 10.4
Arizona 9.7 7.3
Arkansas 13.1 4.2
California 13.3 10.7
Colorado 10.5 9.8
Connecticut 12.2 12.6
Delaware 14.8 *12.7
Florida 16.2 **
Georgia 13.3 10.1
Hawaii 8.6 *4.4
Idaho 9.3 8.2
Illinois 15.2 5.8
Indiana 15.6 **
Iowa 14.8 *10.3
Kansas 18.6 **
Kentucky 12.5 *10.6
Louisiana 22.7 *6.7
Maine 14.2 11.7
Maryland 13.4 10.9
Massachusetts 17.9 16.2
Michigan 14.8 5.5
Minnesota 18.2 *6.5
Mississippi 16.9 **
Missouri 12.9 8.1
Montana 8.0 7.9
Nebraska 12.2 8.2
Nevada 9.0 7.4
New Hampshire 13.2 10.8
New Jersey 14.1 12.7
New Mexico 9.1 8.7
New York 15.7 6.8
North Carolina 15.7 9.7
North Dakota 14.7 *14.5
Ohio 14.2 *3.5
Oklahoma 16.5 **
Oregon 12.1 9.0
Pennsylvania 14.6 11.9
Rhode Island 18.7 17.6
South Carolina 15.8 *4.0
South Dakota 12.9 9.1
Tennessee 14.8 9.1
Texas 17.5 *3.6
Utah 9.9 9.1
Vermont 11.1 8.3
Virginia 12.2 8.1
Washington 14.3 11.5
West Virginia 15.6 10.9
Wisconsin 11.7 *8.6
Wyoming 6.4 6.0
*Estimate based on small sample size.
**Sample size too small to report data reliably.
The geographic distribution in Figure 2
(the average number of days anglers
spent fishing for trout) is somewhat
similar to Figure 1 which shows
freshwater participation rates in trout
fishing. In general, the comparison shows
that states with the highest levels of
trout days also tend to be the states with
the highest levels of trout fishing
participants. However, some Western
states with high levels of participation
had low average days of participation.
For example, Wyoming had a low average
day of trout fishing (6 days) yet the
highest level of trout participation by
freshwater anglers (94 percent). One
explanation may be that anglers in these
states are not avid anglers and therefore
the number of days they fish for trout is
low. This would lower the average for the
state. Average days for all freshwater
fishing is also low for Western states
indicating that non-avidity is not confined
to trout fishing but is true for freshwater
fishing in general in some Western states.
Some states such as North Dakota and
Delaware had high average days but low
levels of angler participation. However,
the sample sizes for these states are less
than 25 which brings into question the
reliability of those estimates.
Freshwater fishing is a very popular
activity with nearly 14 percent of the
U.S. population 16 years of age and older
participating in 1996. Thirty-one percent
of all freshwater anglers fished for trout,
nearly 4 percent of the U.S. population 16
years of age and older. In the following
pages we present a comparison of
freshwater anglers and trout anglers to
the U.S. population by age, sex,
education, income, geographic region,
and population density of residence.
Tables 5 through 10 show the proportion
of the population that participates in
each activity for each category (e.g.,
what proportion of the 45-54-year-old
U.S. population fishes in freshwater and
what proportion of the 45-54-year-old
freshwater anglers trout fishes). The
columns labeled “Percent” in tables 5
through 10 show the percent of
participants in each activity by category
(e.g., what percent of all freshwater
anglers were 45-54 years old). Because
of the relatively large sample sizes for
national estimates, differences in
characteristics that are 3 percent or
larger are usually significant at the
90 percent confidence level.1
Age
Trout fishing appeals to all age groups.
At least 29 percent of all freshwater
anglers in all age categories fished for
trout (Table 5). However, most trout
anglers fall between the ages of 25 and
54 (68 percent) with people between
the ages of 35 and 44 (27 percent)
comprising the single largest cohort of
trout anglers.
Comparing trout anglers to the U.S.
population shows that trout anglers were
younger than the general population.
Fifty percent of trout anglers were
between 25 and 44 while only 41 percent
of the general population were in this age
group. The percent of trout anglers 55
and older (17 percent) is lower than that
of the U.S. population in that category
(28 percent). Trout anglers and
freshwater anglers had a similar age
distribution.
Sex
Seventy-eight percent of trout anglers
were male. This is disproportionately
high compared to the U.S. population,
where women were the majority at 52
percent (Table 6). The percent of male
trout anglers (78 percent) was also
higher than the percent of all male
freshwater anglers (74 percent).
While many females 16 years of age and
older participated in freshwater fishing
(7.6 million), this was only 7 percent of
6
Characteristics of
Trout Anglers
Table 5. Age Distribution of the U.S. Population, Freshwater Anglers, and Trout Anglers: 1996
(Population 16 years of age and older. Numbers in thousands. Excludes Great Lakes fishing.)
Freshwater Anglers Trout Anglers
U.S. Population Percent Percent of
of U.S. Freshwater
Age Number Percent Number Percent Population Number Percent Anglers
U.S. Total 201,472 100 28,921 100 14 8,974 100 31
16-17 7,222 4 1,224 4 17 369 4 31
18-24 25,120 12 3,493 12 14 1,014 11 29
25-34 40,918 20 7,037 24 17 2,047 23 29
35-44 42,600 21 7,632 26 18 2,443 27 32
45-54 31,204 15 4,806 17 15 1,579 18 33
55-64 21,213 11 2,610 9 12 831 9 32
65+ 33,670 17 2,399 8 7 716 8 30
Note: Data may differ from previous reports due to ratio adjustments of age cohorts. Detail does not add to total due to rounding.
Table 6. Sex Distribution of the U.S. Population, Freshwater Anglers, and Trout Anglers: 1996
(Population 16 years of age and older. Numbers in thousands. Excludes Great Lakes fishing.)
Freshwater Anglers Trout Anglers
U.S. Population Percent Percent of
of U.S. Freshwater
Sex Number Percent Number Percent Population Number Percent Anglers
U.S. Total 201,472 100 28,921 100 14 8,974 100 31
Male 96,660 48 21,371 74 22 7,045 78 33
Female 104,812 52 7,550 26 7 1,929 22 26
1 This means that for 90 percent of all possible
samples, percentage differences 3 percent or
greater are reliable indicators of differences.
the female population in the United
States. In comparison, 22 percent of the
male population 16 years of age and older
participated in freshwater fishing. In
addition, female anglers were less likely
to participate in trout fishing than male
anglers. Female freshwater anglers
participated in trout fishing at a rate of
26 percent while male freshwater anglers
participated at a rate of 33 percent. As a
result, females made up 26 percent of
freshwater anglers and 22 percent of
trout anglers.
Education
People of all educational levels
participated in trout fishing although
trout anglers tend to have completed
more years of education than the U.S.
population and than all freshwater
anglers. Table 7 shows that 17 percent
of Americans 16 years of age and older
had not completed high school. Only
11 percent of trout anglers had not
completed high school. Fourteen percent
of all freshwater anglers fell into this
category. A higher proportion of trout
anglers had 4 years or more of college
(31 percent) compared to the U.S.
population (25 percent). Twenty-six
percent of all freshwater anglers had 4
years or more of college.
Freshwater anglers’ participation in trout
fishing increased with a rise in the level
of education — from 25 percent for
anglers with less than 12 years of
education to 37 percent for anglers
with 4 or more years of college.
Income
In 1996, the median household income for
the U.S. was slightly more than $35,000.
Freshwater anglers had higher incomes
than the U.S. population. Fifty-five
percent lived in households with annual
incomes $35,000 and greater (Table 8).For
trout anglers, an even greater proportion
(59 percent) came from households with
incomes above the national median
income. Households with incomes below
the median accounted for 33 percent of
all freshwater anglers and 30 percent of
trout anglers. Twelve percent of all
freshwater anglers and 11 percent of
trout anglers lived in households where
income was not reported.
At least 25 percent of all freshwater
anglers in each income category fished
for trout. Freshwater anglers from
households earning less than $10,000 a
year participated in trout fishing at a rate
of 25 percent. Freshwater anglers from
households earning more than $100,000
had the highest participation rate, 38
percent. The next highest participation
rate for trout fishing by freshwater
anglers is the $75,000-$99,900 income
category with 36 percent. Freshwater
anglers from homes not reporting their
income participated in trout fishing at a
rate of 29 percent. Nineteen percent of
all respondents did not report their
household income.
7
Table 7. Educational Distribution of the U.S. Population, Freshwater Anglers, and Trout Anglers: 1996
(Population 16 years of age and older. Numbers in thousands. Excludes Great Lakes fishing).
Freshwater Anglers Trout Anglers
U.S. Population Percent Percent of
of U.S. Freshwater
Education Number Percent Number Percent Population Number Percent Anglers
U.S. Total 201,472 100 28,921 100 14 8,974 100 31
0-11 Years 33,743 17 3,959 14 21 981 11 25
12 Years 71,098 35 10,470 36 15 3,079 34 29
1-3 Years College 45,573 23 7,017 24 15 2,144 24 31
4+ Years College 51,057 25 7,474 26 29 2,770 31 37
Table 8. Income Distribution of the U.S. Population, Freshwater Anglers, and Trout Anglers: 1996
(Population 16 years of age and older. Numbers in thousands. Excludes Great Lakes fishing).
Freshwater Anglers Trout Anglers
U.S. Population Percent Percent of
of U.S. Freshwater
Household Income Number Percent Number Percent Population Number Percent Anglers
U.S. Total 201,472 100 28,921 100 14 8,974 100 31
Under $10,000 15,043 7 1,164 4 8 291 3 25
$10-19,900 19,690 10 2,199 8 11 602 7 27
$20-24,900 13,080 6 1,852 6 14 521 6 28
$25-29,900 12,337 6 2,287 8 19 591 7 26
$30-34,900 12,572 6 2,054 7 16 616 7 30
$35-39,900 10,653 5 1,999 7 19 587 6 29
$40-49,900 18,274 9 3,581 12 20 1,150 13 32
$50-74,900 32,223 16 5,988 21 19 1,997 22 33
$75-99,900 28,835 7 2,444 8 16 868 10 36
$100,000 or More 13,756 7 1,968 7 14 756 8 38
Not Reported 38,774 19 3,394 12 9 994 11 29
Census Geographic Region
Table 9 shows the distribution of all
freshwater and trout anglers by Census
Geographic Divisions. These geographic
regions are defined in Figure 3. As the
state-by-state analysis suggests, the
regions with the greatest share of trout
anglers are the Mountain (21 percent)
and Pacific (29 percent) regions. This
geographical distribution is quite
different from that of total freshwater
anglers. Only 8 percent of total
freshwater anglers live in the Mountain
region and 12 percent in the Pacific.
Also, although the East North Central
region has one of the largest shares of
freshwater anglers (17 percent), only
6 percent of trout anglers live in this
region.
Overall, 31 percent of all freshwater
anglers fished for trout. The percentage
of freshwater anglers participating in
trout fishing by region varied from a low
of 9 percent in the East South Central to
highs of 81 and 72 percent in the
Mountain and Pacific regions
respectively. The New England and
Middle Atlantic regions also had high
levels of participation (61 and 51 percent,
respectively).
Population Density of Residence
The 1996 FHWAR asked respondents
whether they considered their place of
residence to be a big city or urban area,
a small city or town, or a rural area.
These categories were not defined for
the respondent (e.g., by big city we mean
“a city with a population of 500,000 or
more”). Consequently, one respondent
may consider an area to be a small city
while another resident may consider the
same area a big city. Therefore, the
results discussed below should be viewed
from the perspective of where the
respondents classified themselves as
living and not some generally assigned
definition for the size of a big city, small
city or rural area.
As a percent of the U.S. population,
people in rural areas participated in
freshwater fishing almost twice as much
as residents of big cities, 21 percent
versus 11 percent (Table 10). As a percent
of total freshwater anglers, the largest
number of participants are residents of
small cities and towns (41 percent).
Similarly, as a percent of total trout
anglers, most trout anglers are from
small cities and towns (42 percent).
8
Figure 3. Map of U.S. Showing Bureau of Census Regions
West
Mountain West
North Central
West
South
Central
East
South
Central
South
Atlantic
East
North Central
Middle
Atlantic
New
England
Pacific
Midwest
South
Northeast
AK
Alaska
WA
Washington
ID
Idaho
OR
Oregon
CA
California
HI
Hawaii
NV
Nevada
UT
Utah
AZ
Arizona NM
New Mexico
CO
Colorado
WY
Wyoming
MT
Montana
ND
North
Dakota
MN
Minnesota
IA
Iowa
WI
Wisconsin
OH
Ohio
KY
Kentucky
WV
VA
Virginia
TN
Tennessee
GA
Georgia
FL
SC
IL
Illinois
MO
Missouri
SD
South
Dakota
NE
Nebraska
KS
Kansas
OK
Oklahoma
AR
Arkansas
TX
Texas
AL
Alabama
IN
Indiana
LA
Louisiana
MS
Mississippi South
Carolina
NC
North Carolina
Florida
West Virginia
DE Delaware
NJ New Jersey
CT Connecticut
RI Rhode Island
MA Massachusetts
NH New Hampshire
VT Vermont
MD Maryland
PA
Pennsylvania
ME
Maine
NY
MI New York
Michigan
9
Table 9. Geographic Distribution of the U.S. Population, Freshwater Anglers, and Trout Anglers: 1996
(Population 16 years of age and older. Numbers in thousands. Excludes Great Lakes fishing.)
Freshwater Anglers Trout Anglers
U.S. Population Percent Percent of
of U.S. Freshwater
Region Number Percent Number Percent Population Number Percent Anglers
U.S. Total 201,472 100 28,921 100 14 8,974 100 31
New England 10,306 5 1,143 4 11 694 8 61
Middle Atlantic 29,371 15 2,529 9 9 1,289 14 51
E. North Central 33,121 16 4,963 17 15 531 6 11
W. North Central 13,875 7 3,244 11 23 464 5 14
South Atlantic 36,776 18 4,774 17 13 942 10 20
E. South Central 12,459 6 2,339 8 19 199 2 9
W. South Central 21,811 11 4,046 14 19 417 5 10
Mountain 11,966 6 2,283 8 19 1,854 21 81
Pacific 31,787 16 3,599 12 11 2,584 29 72
Table 10. Population Density Distribution of the U.S. Population, Freshwater Anglers, and Trout Anglers: 1996
(Population 16 years of age and older. Numbers in thousands. Excludes Great Lakes fishing.)
Freshwater Anglers Trout Anglers
U.S. Population Percent Percent of
of U.S. Freshwater
Density of Location Number Percent Number Percent Population Number Percent Anglers
U.S. Total 201,472 100 28,921 100 14 8,974 100 31
Big City/Urban 73,480 36 8,013 28 11 2,680 30 33
Small City/Town 83,720 42 11,978 41 14 3,736 42 31
Rural 42,976 21 8,883 31 21 2,555 28 29
No Response 1,295 1 47 0 4 3 0 6
From 1991 to 1996 the estimated number
of freshwater anglers changed slightly
from 30.2 million to 28.9 million.
Similarly, the number of trout anglers
remained almost the same — 9.1 million
in 1991 and 9.0 million in 1996. Both of
these changes are within the margin of
error for the estimates at the 95 percent
confidence level so we cannot be sure that
the number of anglers actually
decreased.2 However, fishing activity as
measured by fishing days has increased.
All freshwater fishing days rose 13
percent and trout fishing days increased
7 percent. Corresponding with this, the
average fishing days per angler increased
for all freshwater anglers from 14 days in
1991 to 17 days in 1996, and for trout
anglers from 9 days in 1991 to 10 days in
1996. This indicates that all freshwater
and trout anglers are not increasing in
number of participants but that they
are more avid, that is, they spend more
time fishing per person.
One explanation for the increase in the
number of days is that the average
angler was older in 1996 than he was in
1991 and consequently may have had
more leisure time. As shown in Figure 5,
the numbers of all freshwater anglers
between the ages of 35 and 54 increased
from 36 percent in 1991 to 43 percent in
1996. Trout anglers between these ages
increased from 37 percent in 1991 to 45
percent in 1996. The aging of anglers is
related to the aging of the “Baby Boom”
generation. The U.S. population between
the ages of 35 and 54 increased from 33
percent in 1991 to 37 percent in 1996.
10
1991–1996 Comparison of
Trout Fishing Activity
Figure 4. 1991-1996 Participants and Days
(In millions)
* The difference is within the margin of error of the estimates at the 0.05 level of significance. This means
that for 95 percent of all possible samples, the estimates for the number of freshwater and trout anglers in
1991 are not different from the respective estimates for 1996.
2 This means that for 95 percent of all possible
samples, the estimates for the number of all
freshwater and trout anglers in 1991 are not
different from the estimates for 1996.
30.2
9.1 9.0
485
431
93.6
81.4
Number of Trout Anglers*
10
8
6
4
2
0
28.9
Number of Freshwater Anglers*
35
30
25
20
15
10
5
0
Freshwater Days of Fishing (+13%)
500
400
300
200
100
0
Trout Days of Fishing (+7%)
100
80
60
40
20
0
1991 1996 1991 1996
1991 1996 1991 1996
Table 11. 1991-1996 Comparison of Activity: Participants and Days of Fishing
(Population 16 years of age and older. Numbers in thousands.
Excludes Great Lakes fishing.)
1991 1996 Percent
Participants and Days Number Number Change
Anglers, All Freshwater 30,186 28,921 * – 4.0
Anglers, Trout 9,107 8,974 * – 1.5
Days, All Freshwater 430,922 485,474 13.0
Days, Trout 81,366 93,566 7.0
11
Note: Data may differ from previous reports due to ratio adjustments for age cohorts.
Detail does not add to total due to rounding.
Figure 5. 1991-1996 Comparison of Age Distribution of the U.S. Population,
Freshwater Anglers and Trout Anglers
(Population 16 years of age and older. Excludes Great Lakes fishing.)
1991 1996
1991 1996
1991 1996
Percent of U.S. Population, by Age
Percent
30
25
20
15
10
5
0
Age 16-17 18-24 25-34 35-44 45-54 55-64 65 plus
Percent of Freshwater Anglers, by Age
Percent
30
25
20
15
10
5
0
Age 16-17 18-24 25-34 35-44 45-54 55-64 65 plus
Percent of Trout Anglers, by Age
Percent
30
25
20
15
10
5
0
Age 16-17 18-24 25-34 35-44 45-54 55-64 65 plus
12
The descriptive statistics presented in
the previous section show that trout
anglers are different from all freshwater
anglers in some ways. These descriptive
characterizations of anglers have
limitations. First, without conducting the
appropriate statistical test, it is
impossible to determine whether an
observed difference between two groups
is statistically significant. However,
because of the relatively large sample
sizes for national estimates, differences
in characteristics that are 3 percent or
larger are usually significant at the 90
percent confidence level. Second, even if
the difference was statistically
significant, the isolated effect of the
characteristic on an angler’s decision to
fish for trout cannot be measured. For
example, in the general population,
income level is correlated with gender.
The previous section found participation
in trout fishing is also correlated with
gender. This raises the question: Were
men more likely to have fished for trout
because of their gender or because they
were more likely than women to come
from households with higher income
levels? A participation model may be
used to analyze this type of question.
The probability of fishing for trout was
estimated to predict what sort of angler
was most likely to fish for trout and to
evaluate the isolated effects of
sociodemographic and other factors on
that decision. In participation models,
the effect of a particular characteristic
is calculated in an “other things being
equal” context. In the example above,
this procedure removes the confounding
effects of the correlation between
gender and income to show how each
characteristic contributes to participation
in trout fishing.
The model hypothesizes that a
freshwater angler’s decision whether or
not to fish for trout, given that he or she
already fishes in freshwater, depends on
the angler’s sociodemographic
characteristics, whether the angler fished
more than average, the region of the
country where he or she resides, and the
water quality (WQ) of rivers and streams
in the respondent’s state of residence.
The RiverWQ variable represents the
percentage of river and stream shore
miles described by the EPA as “good
water quality that fully supports aquatic
life uses.” Data for this variable were
taken from the Environmental Protection
Agency’s 1996 National Water Quality
Inventory biennial report to Congress.
These data are based on information
collected and evaluated by the states in
1994 and 1995.3
The participation variable is one if the
angler fished for trout and zero if he or
she did not fish for trout. This type of yes
or no response is modeled in the
logarithm of the odds that the individual
fished for trout. This is called the logit
equation and appears on the left side of
equation 1. Equation 1 shows the model
estimated.
where:
Pi= Probability that the i-th individual
fished for trout
xi= Vector of explanatory variables
α = Intercept to be estimated
β = Vector of coefficients to be estimated
The explanatory variables, xi, were a
combination of binary and continuous
(1)
Pi ln _____ = α + βxi (l-Pi)
Participation Model
Table 12. Explanatory Variables in the Trout Fishing Model
Variable Description Mean
Income Annual household income, 49.5
in thousands of dollars
Urban 1 if respondent indicates urban residence 0.28
0 otherwise
Female 1 if respondent is female 0.26
0 otherwise
Avid 1 if respondent fished 20 days or more 0.22
0 otherwise
African-American 1 if respondent’s ethnicity is African-American 0.05
0 otherwise
Hispanic 1 if respondent’s ethnicity is Hispanic 0.03
0 otherwise
Northeast 1 if respondent resided in the Northeast4 0.39
0 otherwise
South 1 if respondent resided in the South5 0.39
0 otherwise
West 1 if respondent resided in the West6 0.20
0 otherwise
RiverWQ Percent of river and stream miles reported to have 55.7
good water quality that fully supports aquatic life.7
4 Northeast includes Middle Atlantic and New England regions as defined by the U.S. Census Bureau.
See Figure 3.
5 South includes the West South Central, East South Central, and South Atlantic regions as defined by the
U.S. Census Bureau. See Figure 3.
6 West includes Mountain and Pacific regions as defined by the U.S. Census Bureau. See Figure 3.
7 Data from EPA’s National Water Quality Inventory: 1996 Report to Congress.
3 States do not use identical survey methods and
criteria to rate their water quality. There are no
data available for Alaska, Oregon and Idaho.
13
The equation also shows that the
probability of anglers fishing for trout
increases with the water quality of rivers.
This result makes intuitive sense as trout
fishing would be more desirable when
water quality is good.
The estimated coefficients do not
provide a direct measure of how the
explanatory variables affect the
probability that a freshwater angler will
fish for trout. The coefficients show the
effect of the variable on the logarithm
of the odds ratio. To get around this,
partial derivatives were calculated at the
means of the continuous variables with
all dichotomous choice variables equal to
zero to show how each variable affects
the probability of fishing for trout. The
partial derivatives shown in Table 13 can
be used to make statements like “being
a resident of a Western state increases
the probability of fishing by 40 percent.”
The partial derivatives in Table 14 show
a change from a base case of a non-
African-American, non-Hispanic male,
with average income who lives in a
Midwestern state with average river
water quality.
The whole equation can also be
reevaluated to make more complex
comparisons. For example, a non-urban,
male, African-American angler residing
in a Southern state has a 3 percent
chance of fishing for trout, while an
urban, avid, Hispanic male angler
residing in the West has a 91 percent
chance of fishing for trout. Because the
model predicts the log of the odds ratio,
the calculation of these probabilities is
variables. They are described in Table 12.
The means of the binary variables repeat
some of the percentages reported earlier.
For example, 26 percent of freshwater
anglers are women. The region of
residence provides a rough indicator of
the availability and quality of trout
fishing sites.
Table 13 shows the model estimated from
a nationwide sample of 7,984 freshwater
anglers. All variables were significant at
the 1 percent level except income which
was significant at the 5 percent level.
Variables for age and retired anglers
initially included in the model were
excluded after chi-square likelihood ratio
tests determined that their coefficients
were not significantly different from zero
and therefore did not provide additional
predictive power to the model.
The likelihood ratio index can be
interpreted in a similar way as a multiple
correlation coefficient (R2) in ordinary
least squares regression. The index value
of 0.15 indicates that the equation
explains about 15 percent of variation in
the logit, which is typical of recreation
models. The equation shows that the
probability of fishing for trout increases
with income and urban residence, other
things being equal. It also shows that
anglers who fish 20 days or more, or who
live in the West or Northeast, have a
higher probability of fishing for trout.
Hispanic anglers are more likely to trout
fish. Female anglers, African-American
anglers and anglers who live in the South
were less likely to fish for trout, other
things being equal.
rather involved. A note at the end of this
document explains the calculation.
The results show that gender, race,
region of residence, avidity, income and
water quality have significant impacts
on the probability of trout fishing. The
effects of income and water quality are
not as strong as the other explanatory
variables. For example, a $1,000 increase
in income or a 1 percent increase in river
water quality increased the probability of
participation by less than one percent.
Table 13. Logit Equation Results for Trout Fishing Model
Estimated Standard Partial
Variable Coefficient Error Derivative
Intercept –1.635 0.095 —
Income* 0.002 0.001 0.0004
Urban 0.236 0.061 0.0417
Female –0.220 0.064 –0.0389
African-American –1.067 0.165 –0.1889
Hispanic 0.729 0.159 0.1290
Avid 0.311 0.064 0.0551
South –1.170 0.107 –0.2071
West 2.236 0.091 0.3958
Northeast 1.152 0.087 0.2040
RiverWQ 0.006 0.001 0.0010
Pseudo R2 = 0.153
All variables significant at the 0.01 level.
* in thousands
2. Multiply each value by the beta for
that variable. The betas are the
estimated coefficients in Table 13.
3. Sum the results in the calculation in
step 2. In this example, the sum is 2.2026.
This is the α+βx term from equation 2.
4. Take the antilog of the sum. On most
calculators this is the ex function. In most
spreadsheets it is the @EXP (.) function.
This is 10.0002 in this example.
5. Calculate Pi as the result of step 4
divided by the sum of 1 and the result of
step 4 as shown in equation 2, 0.9091 in
this example.
The probability is interpreted as the
probability that an angler with the
characteristics described by the
independent variables will be a trout
angler. In this case, the probability is
90.91 percent that an urban, avid,
Hispanic male who resides in a western
state is a trout angler.
14
Note on Calculation Probabilities of
Participation from the Trout
Participation Model
Equation 1 (page 12) can be solved for the
individual’s probability of participation in
trout fishing, Pi, as
As they appear in both the numerator
and the denominator of this equation, the
interpretation of the coefficients (or
“Betas”) shown in equation 2 is not
direct. An individual’s probability of
participation can be calculated using the
coefficients in equation 2 and a scientific
calculator or a spreadsheet. Table 15
illustrates the process for an urban, avid,
male, Hispanic angler from the West.
1. First, determine the values you wish to
use for each independent variable. The
intercept is α and is constant. For the
example, we assume average income, 49.5
(in thousands) and average water quality
of rivers (55.7 percent). The rest of the
variables are dichotomous, i.e., equal to
one if the statement is true and 0 if it is
false. Notice that Northeast – South –
West, and African-American – Hispanic
are mutually exclusive, if one is true the
other cannot be true.
(2)
eα + βxi
Pi = __________
(l+eα + βxi)
Table 14. Calculation of Trout Fishing for an Urban, Avid, Male, Hispanic Angler from
the West
Variable Beta Value Beta X Value
Intercept –1.6347 1 –1.6347
Income* 0.00231 49.5 0.11434
Urban 0.236 1 0.2358
Female –0.2196 0 0
African-American –1.0668 0 0
Hispanic 0.7289 1 0.7289
Avid 0.3111 1 0.3111
South –1.1696 0 0
West 2.2358 1 2.2358
Northeast 1.1522 0 0
RiverWQ 0.00559 55.7 0.311363
Sum (α+βx) 2.2026
e(α+βx) 10.0002
Probability 0.9091
* in thousands
15
The 1996 National Survey of Fishing,
Hunting, andWildlife-Associated
Recreation shows that fishing for trout
appealed to a large number of freshwater
anglers. From a sociodemographic
standpoint, there are some interesting
differences between freshwater anglers
overall and trout anglers. Trout anglers
tend to be male, have higher incomes, and
live inWestern or Northeastern regions.
A trout fishing probability of participation
model showed that income, region of
residence, population density of residence,
gender, race and water quality had
significant impacts on whether an angler
sought trout or another species. Also of
note is the finding that avid freshwater
anglers are more likely to fish for trout
than non-avid freshwater anglers. This
means that trout anglers tend to be more
avid than freshwater anglers in general.
These findings from the 1996 FHWAR
underscore the importance of trout for
millions of freshwater anglers.
Information about whom these trout
anglers are, their age, sex, race, where
they live, and so on can be used by
recreation managers and others to
develop and refine fishing management
programs so that anglers’ fishing
experiences can be enhanced.
Summary
© PhotoDisc
U.S. Department of the Interior
U.S. Fish & Wildlife Service
Division of Economics
Arlington, Virginia
www.fws.gov